Thein

[@levelsio @levelsio]

Open an IBKR account, then buy:

**S&P500**
If US citizen: buy VOO
If non-US citizen: buy VUAA on LSEETF (search VUAA on IBKR)

**Tech**
If you want more tech exposure (more gains but more risk), if US buy QQQ, if non-US buy XUTC

**Bonds**
If you want to keep some money safe put it in bonds, but that also has an ETF, if US buy SGOV, if non-US buy IB01

**Commodities** (e.g. materials)
Want gold? If US buy IAU, non-US by RMAU

**Why difference US vs non-US citizen investing?**
It's cheaper: Irish-domiciled UCITS ETFs generally pay 15% US withholding tax on dividends from US companies, compared to 30% for funds without equivalent treaty access

The only proven thing in investment is that diversification works, you have less gains but less risk, ask AI how to build a nice mix of ETFs, bonds, commodities, real estate/land (p.s. ETFs are categorized as stocks cause they're just a basket of stocks), generally you want to not buy individual stocks with most of your money because it's super volatile and you're betting. It can go both ways, I bet NVIDIA in 2022 and made lots of money but I also bet Alibaba before Jack Ma was arrested and lost 90%!

This is not financial advice

[ClaudeDevs @ClaudeDevs]

https://t.co/v694m8Eaj6

[Owen Carey @owenthcarey]

Practicing LeetCode at 32 years old.

[𝐃𝐚𝐫𝐤 𝐇𝐚𝐢𝐫𝐞𝐝 @DKH013]

Proof that 99% of what we worry about never actually happens.

[Thein @realthein]

@Thenationth @grok is it true?

[Thein @realthein]

@seraleev Oh right. Kill the app and reopen. working already

[Thein @realthein]

@EpicNewsroom @grok how many affected ?

[Thein @realthein]

@Keir_Starmer @grok summary

[NeetCode @neetcode1]

We really really need more people who think deeply about things.

I've always felt this way, but even more so now in the age of AI.

In school, most people are just looking to complete their assignments, get their code working, and then call it a day. I know I definitely did.

In the professional world, you have to be more careful, but the exact same pattern emerges. A feature is shipped, it satisfies the most basic requirements, and you're done.

But is it actually good? Is it useful? Did you really produce anything of value? Or are you just trying to check a box.

Not enough people are asking these questions, and that includes the biggest of tech companies - not to pick on Microsoft, but they shipped Github Copilot very early, and then later MS Copilot. Fantastic ideas, but they fell short, and that's the only reason a company like Cursor was able to exist in the first place.

And now, in 2026, it's never been easier to build things. But I would say that just makes it 10x harder to actually build value.

If you take the extra time to slow down and actually refine a solution, you unlock exponentially more value. Less and less people are doing this for some reason - well I think we know the reason, but still..

and for the record, I'm not saying I haven't ever fallen into this trap, but if I ever had any success in life it was only because I spent a little more time thinking about certain things, that most people overlooked. When I look at the most successful people, I think this is a consistent pattern that emerges

[Thein @realthein]

@grok @AIOptimist1 @AravSrinivas @grok give me one best example usage of that ?

[Thein @realthein]

@DataChaz Oh that might explain even I used so much, quota is dropped only a bit.
@grok when it started and when it will finish

[Thein @realthein]

Switched from Chrome to Atlas after getting ChatGPT. feels like OpenAI is building a full ecosystem.

[Thein @realthein]

Switched to Codex Pro and confirmed it works with OpenClaw too.

Claude Pro is cheaper, but the quota disappears fast. usually gone after 2–3 complex tasks on my side projects.

[Thein @realthein]

@TheRealAdamG @steipete @grok can open claw use with ChatGPT Plus subscription?

[Thein @realthein]

@bindra_dhruv @grok is it true?

[Thein @realthein]

@MrinankSharma @grok summarise

[Thein @realthein]

@MKBHD @Snapdragon @grok what’s the situation now ?

[Uncle Bob Martin @unclebobmartin]

Brilliant!

“The Confidence Spiral: The more AI writes, the less you trust your own judgment. The less you trust your judgment, the more you defer to AI. The more you defer, the less you learn. The less you learn, the less you trust yourself. Spiral continues.”

[Boris Cherny @bcherny]

I'm Boris and I created Claude Code. I wanted to quickly share a few tips for using Claude Code, sourced directly from the Claude Code team. The way the team uses Claude is different than how I use it. Remember: there is no one right way to use Claude Code -- everyones' setup is different. You should experiment to see what works for you!

[Charly Wargnier @DataChaz]

folks on X: "Clawdbot is an overnight success"

@steipete’s GitHub profile: